The Luminous Ambience of an Ancient Roman Public Building: A Characterization of the Inner Daylit Environment of Rogatianus Library in Thamugadi City (Timgad, Algeria)

Abstract

The Roman public library of Timgad (Algeria) constituted the study object of several extensive research works, particularly during the French colonial era. Following a virtual restitution-focused research work, this investigation aims to quantitatively characterize the daylighting conditions inside Timgad Public Library. Here, it must be remembered that the luminous environment inside libraries is a main design parameter and a main environmentalfactor. In addition, it must be highlighted that the Timgad region’s luminous climate differs from where Rome’s designers and builders practiced. Hence, at a first step, a comparison is carried out between the precepts of Vitruvius, the pioneer of ancient Roman architecture, and the outcomes of previous studies related to Roman libraries. Then, as a second step, a double approach combining both 3D geometric modeling and numerical simulation using Radiance (2.0Beta) software. These simulations are mainly elaborated for the case of the large reading room, including most of the building’s main activities—consulting, reading, and storing books. Finally, this inner luminous environment’s characterization highlights that the Timgad Roman antic library was not uniformly daylit and suggests that its use varied spatially and temporally with respect to this environmental parameter.

1. Introduction

The ancient world knew the building of great libraries such as those of Alexandria in Egypt and Pergamum in Asia Minor. These libraries constituted a design reference for Roman builders who took up certain earlier architectural components for the construction of their libraries [1]. During the Roman Empire era, and throughout the Mediterranean basin, some libraries were founded as a personal gift, of a cultural nature, from the citizen to his city [2]. These monumental buildings were erected in order to commemorate the Founder’s name and works. Such euergetism has given rise to numerous cultural buildings, such as libraries [1]. During this period, libraries experienced a real development in terms of architectural design consisting of the following stages: (i) first, modest places with a cultural vocation, then (ii) inserted within other cultural institutions, and finally (iii) distinct buildings increasingly monumentally expressed during the Roman era [3].

Since their first appearances, the major function of these Roman buildings has been “memorial”. These “memory repositories” [4] were devoted to preserving the official documents and public and/or private archives of the city. However, in certain cases, its memorial function is more personal or dynastic, related to its donor and his family [5]. Furthermore, it also provided an encyclopedic function for the private collections of classical Greece and Republican Rome. Additionally, the library was a conservatory of human intellectual heritage as well as a workplace [6].

As a case study, the library of ancient Rome has faced numerous questions due to several relevant aspects remaining unknown until today. Over the last century, some researchers, such as CH. Callmer (1944) [7], E. Makowiecka (1978) [8] and VM.U Strocka (1981) [9] have attempted to study this type of construction in order to better understand the architecture and management of these buildings. This theme is still relevant in this century and has been the subject of recent research studies carried out, among others, by L. Casson (2002) [10], M. Nickolls (2013) [11], M. Mamoli (2014) [12], G. W. Houston (2014) [13].

In addition, natural lighting was a major concern in the design of public buildings, particularly libraries in Antiquity. During this period, light played a functional, symbolic, and aesthetic role in architecture. The Greeks and Romans used natural light through elements such as the oculus of the Pantheon, and relied on artificial lighting with torches, oil lamps, and candles in temples, amphitheatres, and houses [14]. Light possessed significant symbolic value, particularly in religious contexts, where it was seen as a manifestation of the divine [15]. However, natural lighting in Antiquity remained little studied from a quantitative perspective because many materials used in Roman times, including glass and wood, have disappeared over the centuries. It is now challenging to reconstruct with precision the lighting environment of Roman public buildings.

In library context, C. Balocco and R. Calzolari indicate that the lack of artificial lighting, natural light in the ancient Italian library showed an irregular distribution throughout the day; it peaked around midday, while it was significantly lower in the early morning and late afternoon [16].

North Africa was the fertile ground of distant human occupations and old civilizations, leaving so many remains testifying to the incredible ingenuity of its builders. The Roman civilization is still marking its existence through the presence of many magnificent remains clearly seen in place. Timgad city contains an immense archaeological wealth illustrating the presence of the Roman urban civilization within the Aurès massif in Algeria. Apart from Rome city, it is in Timgad that the unique multifunctional library in this western part of the Roman Empire is located [17].

Timgad Library was considered a learning and cultural center for the city inhabitants, enhancing education, literacy and intellectual development [4]. Thus, its presence as a building informs about this Roman city’s rank as a unique cultural place and intellectual home in the western region of North Africa. Hence, in this research work, Timgad public library is investigated as a North African case study distinct from that of Alexandria, one and because of its notable testimony of the Roman Empire’s cultural achievements in this region.

The present study aims to quantitatively characterize the natural lighting conditions inside the main reading room of the ancient public library of Timgad. Beyond highlighting the distribution of natural light as the main visual source, it attempts to identify the different activities practiced inside the various parts of this building.

2. Natural Lighting in Ancient Roman Public Libraries

Inside the reading spaces like libraries, natural lighting design is both a crucial design parameter and a functional comfort indicator. In fact, the layout of the libraries of the Forum of Trajan and Palatine in Rome, Celsus in Ephesus, as well as Rogatianus in Timgad, reveals the particular use of wide doors. This will provide, among other functions, a wide admission of natural lighting that will allow a good achievement of the reading tasks by the readers [11]. Yet, very few studies have emphasized its importance during the Antiquity era. In fact, it must be noticed that the rooms were lit solely by openings in the main facade for the cases of Classical and Hellenistic [12]. The book storage rooms lacked openings, as the windows were not yet glazed. The windows, closed only by wooden shutters, did not adequately protect the manuscripts from rain and humidity [12]. Therefore, only the entrance wall had doors and windows were protected by porticos. Besides, it must be noticed that both natural light and fire were the available means used within ancient libraries as illuminating devices [5].

Around 100 AD, the use of glass panels in windows became widespread, resulting in a row of windows appearing along the side walls, above the row of niches housing the books. In Roman libraries, windows were placed either in the entrance wall of the main hall, as in the Library of Celsus, or above the niches, as is the casewith the Library of Celsus in Ephesus and the Melitine library in Pergamon (Figure 1) [12].

Moreover, regarding their opening hours, it seems certain that libraries were open exclusively during the day. More specifically, a unique and well-preserved testimony from Antiquity provides information on the number of hours a library was open. An inscription from the Library of Pantainos in Athens, probably placed above the entrance, states that the library “will be open from the first to the sixth hour.”

Hence, while indicating that the library was open from 6 a.m. to noon, it mainly informs that the building is used when maximum exploitation of sunlight occurs. The rest of the time, the access is closed by a gate or a door, as is the case with the archaeological traces of the Ulpian library or that of Timgad. These use times are consistent with a recommendation by the architect Vitruvius in the 1st century, AD, consisting of orienting the library premises towards the east. On his side, M. K. Affleckconcluded that, inside, the readers perceive the natural light as an important contributor to their visual comfort [5].

About the artificial lighting, E. Makowiecka [8] dismisses the use of open flames in an environment consisting of thousands of highly flammable papyrus scrolls as simply too dangerous. Effectively, there is no reason justifying why public libraries in Rome withartificial lighting devices were not used during cloudy days or short dark winter days. However, the use of such risky devices requires the appropriate and necessary precautions. In addition, the library’s opening hours can really affect the amount of artificial lighting used there [5].

3. The Inner Natural Lighting of the Roman Libraries Following Vitruvius

Vitruvius was a Roman architect and engineer from the 1st century BC, well-known for his architectural treatise, namely “De architectura”. This latter is considered to be one of the first architectural books in history that informs about the principles of Roman architecture.

In chapter VI.4.1 of Book X of “De Architectura”, Vitruvius limited his recommendations about libraries to their orientation towards the rising sun. Vitruvius said that such a solar orientation allows readers a morning consultation as well as avoiding the humidification and exteriorization of books [18]. Conversely, no information is provided about the construction of libraries, their architectural forms, or their spatial organization. However, this orientation-based recommendation should not be considered as a general guideline for the public libraries’ design. In fact, Vitruvius is here referring to the personal libraries inside the private residences that are different from the public institutions’libraries. Such a lack of information about the public libraries is unexpected from Vitruvius. This is particularly noticeable because he dedicated, in his fifth book, an entire section about the types of public buildings within the Roman city walls [12].

Indeed, many famous public libraries’orientations were not respectful to Vitruvius’ recommendations but were rather aligned to their monumental context [11]. Similarly, M. Mamoli argues that, among the seventeen cases in his library’s study corpus, only two cases, the Celsus and the Philippi ones, were respectful to Vitruvius’orientation-related recommendation [12]. Moreover, the well-known Pergamon library is southeast orientated, whilst all the other ones have different and varied orientations.

Hence, one can notice that no concretely defined recommendation was set by Vitruvius that is related to the inner luminous environment design of libraries. Additionally, the inherited ancient libraries did not highlight uniform adoption of such design parameters.

4. Case Study: The Roman Public Library of Timgad

Timgad, the Ancient Roman city Thamugadi, is situated in the North-east of Algeria (Figure 2). It is cited in the Peutinger Table [19] as well as by S. Gsell in his Archeological Atlas of Algeria [20]. Several archeologists worked on the site and revealed what is currently described as one of the best conserved Roman cities throughout the World [21].

Inside this city, the Roman library is located on the northern section of the Cardo, about fifty meters from its crossing with the Decumanus Maximus [22] (Figure 3). According to previous on-site studies, this building had certainly occupied the place of an “insula” or private house [23]. The library was founded and financed by M. Julius Quintianus Flavius Rogatianus, a wealthy citizen from Timgad. He probably died before the library was even completed [13]. The lack of additional information about this donor makes it difficult to set this library’s exact build date. Referring to paleographic bases, R. Cagnat concludes that this library was built by the end of the third century. However, S. Gsell suggests the fourth century instead because of its “dry and thin” capitals. According to him, this architectural detail refers to the Lower Empire [22].

Figure 2. The location of Timgad city [24].

Figure 2. The location of Timgad city [24].

Figure 3. Situation of the Roman public library of Timgad by Google Earth.

Figure 3. Situation of the Roman public library of Timgad by Google Earth.

For six years, the identification of the library was the subject of various controversies among researchers. E. Boeswillwald and R. Cagnat were uncertain about the main function of the building called “semi-circular building,” saying, “the exact destination of the building is unknown. No inscription that tells us, no figurative representation, no characteristic detail.” [22]. Finally, in 1906, three inscription fragments of the building of Timgad allowed it to be identified definitively as a municipal library (Figure 4) [22].

From a functional point of view, Timgad Library ensured the accomplishment of various uses including the reading, verification and archiving of books [25]. To the east of this building are two reading rooms: the main one is in the center, and the secondary one is right next to it (Figure 5).

5. Materials and Methods

This research aims to study the luminous ambience inside a public building built in a Roman province, hence in a natural milieu different from the Northern side of the Mediterranean Sea. Considering the Ambience’s conceptual model [26,27], this investigation is limited to the architectural conformation and the luminous physical signal among this model’s dimensions and did not explore the two other dimensions that are the context and the user. The architectural conformation has been created by means of a 3D model integrating the studied building’s specific geometric dimensions. Then, the physical signal, that is, natural light, has been studied by means of a computer-based simulation process.

This first step took into account previous virtual restitutions of the library and allowed the creation of a new 3D model [28]. What should be remembered is that the lack of information related to a monument that has to be restituted constitutes a major challenge.

Consequently, the process of such restitution faces several methodological problems, particularly with regard to the techniques to be used. In fact, the adopted methods could refer to the reconstruction of the archives and/or visible (existing) components. Differently, some approaches suggest restitution hypotheses based on the established facts and analogies [29].

The 3D restitution models considering the heritage building’s present state and the production of a geometric model remain subject to a process including successive interdependent stages. These latter depend on the nature of the existing data and the aimed type of virtual representations [30].

For the present research work, the virtual restitution processes are conducted through three fundamental steps: (i) the scientific documentation, which includes the literature review, (ii) the field work and finally (iii) the infographic approach (Figure 6).

The data collection phase, as the initial step in the virtual restitution process, was conducted using two complementary approaches: (i) a preliminary historical and documentary investigation concerning the Timgad public library, and (ii) an assessment of its current condition through the application of various survey techniques.

Despite the existence of numerous previous studies focusing on the Timgad library, it was deemed necessary to carry out a comprehensive and updated survey of the entire building. This was achieved with greater precision by employing field-measuring instruments. Such an approach facilitates a direct comparison between the existing historical data and the present physical state of the structure. Ultimately, this will enable us to validate or challenge the hypotheses proposed in earlier research.

In addition to the conventional in situ architectural survey, a topographical survey was conducted with the assistance of a professional topographer. The resulting data enabled the transcription of both altimetric and planimetric information from the field. These elements were then scaled and integrated into a georeferenced plan, which serves as a foundational base for obtaining more precise data on the current state of the site, including altitude, dimensions, and the positioning of the various levels (Figure 7).

For our virtual restitution of the Roman public library of Timgad, we faced a number of problems, such as: (i) the walls’ height, (ii) the openings’ often uncertain location, (iii) the wall coverings, and (iv) the generally deteriorated state of the furniture. This leads us to select the hypothetico-deductive approach among the four logical reasoning inferences. It is the most appropriate way to discuss the archaeological data. In this sense, we can use what is known as “parallel examples” or “analogy”.

Then, this virtually restituted model is imported directly into Radiance (2.0 Beta) from Archicad 16 software. Once the 3D model of the library was integrated, we defined the environmental parameters to be simulated, including: (i) the geographical location and climatic data of the region; (ii) the size, shape, and arrangement of architectural elements that promote the natural light admission, such as windows and doors; (iii) the properties of the interior surface covering materials that influences both diffusion and reflection of the received light (

Table 1. Simulation parameters.

Category	Parameter	Value
Location	Timgad (Algeria)
Climate and sky	Sky type	Clear sky (for a typical sunny day)
	Weather file (EPW)	Banta, Algeria.epw
Rendering settings	Quality preset	High
	Image type	Fish-eye 180°
	Image resolution	1024 × 1024
	Ambient bounces	5
Radiance Parameters	Ambient accuracy	0.1
	Ambient resolution	512
	Limit reflections	6

).

Hence, the daylighting simulation will consist of a succession of “sequences”. These latter are constituted dependently of the presence and arrangement of windows that play a structuring role in the organization of space, in addition to their daylighting. After configuring the simulation options, the calculations are launched. Finally, the outcomes are visualized using the rendering tools provided by Radiance (2.0 Beta).

6. 3D VirtualRestitution of Timgad’s Roman Public Library

The adoption of a hypothetical-deductive approach led to a virtual restitution of our case study, that is, the Timgad city Roman library. It took into consideration the main concerned aspects that are the site, specific architectural devices, spatial components and constructive details.

To achieve this task, a reference is made to some available graphic documents (sketches, plans) relating to Roman architecture, in particular the research works of HF. Pfeiffer (1931) [25], M. Nickolls (2013) [11] and M. Mamoli (2014) [12]. Also, we consider the fact that the Roman provincial libraries often followed the architectural design principles of the capital’s libraries, as stated by several researchers. Hence, our restitution is based on an analogy with some similar Roman libraries located in/and outside of Rome.

For this research work, the computer program Archicad 16 software is used for free. This software allowed us to integrate historical data and graphic documents for a more or less accurate virtual representation of the library as it would have been in the past. The application of Graphisoft Archicad offers various advantages of visualization [31]. With the help of this software, it is possible to check the relationships between spaces, freedom of section between the cover and the wall [32].

As a provincial one, the Timgad Roman library incorporated several similar architectural and decorative features to their Roman counterparts. For example, the library of Timgad (Figure 8) adopts an elongated semi-circular shape, typical of Roman libraries, as the Baths of Trajan’s library in Rome (Figure 9).

Referring to Trajan (Rome) and Hadrian (Athens) libraries as models for Timgad’s one, some Roman architectural components used as daylighting apertures were inserted into this research’s case study. In such a way, the main room should have been illuminated by a zenithal central opening and a lateral one that is the front door (Figure 10).

This illustrates an ideal architectural influence for the preservation and capitalization.

About the roof, and as in other libraries, the interior space should have a vaulted or semi-domed ceiling. This design aimed to impress with the space height and enhance the acoustical conditions of the reading room. As attested by A. Ballu [23] and HF. Pfeiffer [25], the barrel vault was probably made of wood, supported by main plaster beams and decorated with caissons and ribs. The wooden beams were common in Roman architecture for supporting roofs.

Adopting an analogical approach and critically examining the various existing restitutions allowed a hypothetical restitution of the Timgad library in a coherent way. The use of computer graphics tools served to acquire a relatively logical illustration of these hypotheses and to create visual representations of Timgad’s library. These outcomes are structured and organized in such a way that they reflect the library users’ common route. Hence, we generated successive views, from outside towards inside the library, representing all the main spatial components of the Timgad library (Figure 11).

7. Simulation of the Natural Luminous Environment Inside the Rogatianus Roman Library (Timgad)

Simulating the inner luminous environment of a historically virtually restituted building is an increasing practice within the built heritage research field and concerns several civilizations’ affiliated reference buildings, including the Roman Empire ones, such as the Pantheon in Rome [34].

The review of previous research works, consisting of computer-based simulations of historical buildings, reveals a common process involving several steps. These latter have been synthesized and adopted for our case study and will be presented in the following.

7.1. Geo-Climatic Context

Previous research on the physical environment of ancient Roman buildings shows a lack of explicit consideration of climate, probably due to a variation deemed negligible at the time. According to Leveau [35], the climate of North Africa during the Roman period was not fundamentally different from that of today, although it was slightly wetter, particularly in the mountainous areas bordering the desert. S. Gsell [36], based on available written sources, concluded that the climatic conditions of Roman Africa bore strong similarities to those of the Mediterranean at the beginning of the 20th century [37].

Regarding the Aures region, in eastern Algeria, morphoclimatic indicators have highlighted a higher humidity level than that observed today. Between approximately 300 BC and 300 AD, around 1000 BC, the region would have benefited from an “ideal climate”, characterized by less aridity, which made it more favorable to human activities than today [38]. In the Radiance (2.0 Beta) software, we imported the climatic file of the Batna Latitude region 35.55, longitude 6.166667, in order to integrate precise weather data into light simulation.

7.2. Urban Context

The building is located in the central district of the city and close to other major public monuments such as the forum, the theater [11]. This library building is oriented East-West.

7.3. Architectural Configuration of the Library Reading Room

Timgad Roman library’s main room is a large hall combining the functions of a reading, consultation, and storage room [25]. A teacher would bring several students here at a time to consult manuscripts and discuss their contents on site. There were no separate school buildings in the African colonies; teachers, partly paid by the authorities, taught in covered booths near the forum, in porticos, or on the upper floors of private houses [38].

This room measures approximately 15 m wide by 10 m deep [22]. However, Houston reports that this room is only 12 m wide by 10 m deep [13]. The plan of this large semicircular room is preceded by two short side walls offset from the semicircle, each approximately eighteen inches (0.46 m) wide [25]. The personal survey undertaken in situ reveals that this room is 13.28 m wide by 10 m deep (Figure 12). The walls are largely constructed of brick but are faced with white or green marble decorated with moldings and bas-reliefs [39]. As for the decorative elements (columns, steps, entablature), they are constructed of cut stone.

7.4. Building Materials

The flooring, walls, and roofing no longer exist on site. Regarding wood, this perishable material was widely used in storage structures and water systems. Their analysis varies depending on their practical application in timber exploitation and processing, depending on their specific uses: (i) construction timber, (ii) equipment timber (furniture or utensils) [40]. In Timgad, Numidian cedar wood provided excellent construction timber [41]. This building material has a reflection coefficient of 0.19 [42].

In order to define the necessary luminous reflection values specific to the various building materials used during the Roman era in Timgad and its region, an effective on-site measurement campaign has been conducted. Using a light meter, several measurements have been done for certain fragments found: (i) on site, and (ii) in the Timgad and Lambèse museums (Figure 13).

These measurements allow calculating the reflection coefficient of the used building materials in

Table 2. In situ measurements of the reflection coefficients of building materials by H. Djouadi.

Building Materials	Reflection Coefficient	Materials
Brick	0.20–0.30
White limestone	0.55 (0.50–0.60)
White marble	0.83 (0.58–0.85)
Mosaic	0.31
Wood	0.19 (0.15–0.40)	No evidence

. It must be remembered that this indicator is the result of the quotient of the quantity of the natural light reflected by a plane’s surface upon the quantity that it received [34].

7.5. Timgad Public Library’s Morphological and Constructive Characteristics for a Daylighting Study

Considering the studied Public Library building as a variably pierced envelope, the impact of the various materials, in terms of both daylight admission and diffusion, takes into account: (i) the reflection of natural light by the building’s external various surfaces (including the covering of the nearest outside soil and the roof’s one) and then their admission inside throughout the various openings, (ii) the direct admission of daylight and sunlight throughout the various zenithal and lateral openings, (iii) the inside diffusion of admitted daylight due to its reflection by the various horizontal and vertical surfaces (including furniture) of the inner side of the building’s envelope (Figure 14).

7.5.1. The Building’s External Various Surfaces

The areas immediately adjacent to the facades of the Timgad Public Library, including its openings, are located exclusively on the west side of the building. These reflective surfaces are both horizontal and vertical. The horizontal plane, covered with light-colored stone slabs, reflects approximately 25% of the incident solar radiation towards the facade openings, particularly during the morning hours when the solar angle is low. The vertical surface, consisting of light-colored rendered masonry, also contributes to a measurable reflection, estimated at roughly15%.

The library’s roof system is composed of various volumes, allowing for different types of reflections toward the skylights. These reflections can be classified as follows: (i) direct reflections from the inclined surfaces of the roof, (ii) diffuse reflections from interior surfaces (vaults), (iii) secondary reflections generated by adjacent architectural elements (pillars, side walls).

7.5.2. Daylighting Devices

In Timgad’s library, under the portico and facing the main street, is a large door giving access to the semi-circular main room. According to R. Cagnat, this entrance, decorated with engaged columns, was 3.8 m high [22]. In addition, HF. Pfeiffer underlines the use of a bronze grille instead of a wooden door for this room entrance, a device that would allow the air circulation necessary for the books [25]. This solution could be easily done without the risk of rain soaking inside, thanks to the protection provided by the portico. The four rectangular rooms were directly accessible through large openings, which confirms the absence of doors. For Houston [13], the related issue is whether there were really windows in these rooms.

Both of the two large rooms, adjacent to the semicircular room, have very narrow doors. As suggested by M. Mamoli [12], these rooms were used to store books, unless there were some other means of illumination. Furthermore, the same author argues that there is no material or immaterial evidence of the window’s existence in the building.

In summary, for this study, we consider that the west-facing skylights have neither shutters nor sun protection and therefore allow a considerable amount of sunlight to pass directly through. Also, the door located on the west facade is considered a metal grille, and therefore is considered a primary source of natural light for the reading room.

7.5.3. The Inner Surfaces Covering

In the reading room, three categories of interior surface coverings are identified: (i) horizontal (floor and ceiling), (ii) vertical (walls), and furniture elements (Figure 15).

Vertical Surfaces

In the main reading room, several columns are present both outside and inside the semi-circular hall (

Table 3. Construction materials of the vertical plan.

Vertical Plan	Construction Materials
Walls	Brick covered with white and green marble
Columns of the central niche	White marble with spiral fluting
Secondary columns	Stone
Bookcase	Cedar wood

). In the entrance courtyard and the semi-circular hall, the number of columns amounts to twelve. These latter are of Corinthian style with a diameter of 0.45 m and 3.5 m high. The columns’ bases are cubic blocks of white limestone, 0.55 m high, 0.6 m wide, and 0.6 m deep. Also, inside and facing the entrance door, is located the famous niche where the statue of Minerva, the goddess of wisdom and studies, is located. This large niche should have a width of 1.80 m and a height of 1.50 m and be flanked by pilasters with two Corinthian columns of 5 m high [22]. Following our field survey, this niche is framed by two projecting columns of white marble with spiral grooves of 0.65 m.

Horizontal Surfaces

The main hall, paved with blue limestone slabs, is surrounded by a series of podia built of ashlar stone. These poduims measure 0.53 m wide by 0.55 m or 0.60 m deep. The podiums of the main niche are larger than the side ones and measure approximately 0.65 m wide and 0.65 m deep. The distance between the two bases is 1.50 m. The podium of the main hall is preceded by two steps (

Table 4. Construction materials of horizontal plan.

Horizontal Plan	Construction Materials
Floor	Blue limestone slabs
Podium	Blue limestone

). However, previous studies’measurements are slightly different from what we observe at the present time. In fact, the height of the podium steps differs from one step to another. According to our architectural field survey, these steps’ dimensions, respectively from bottom to top, are as follows: (i) 0.27 m depth and height 0.10 m, (ii) 0.28 m depth and0.18 m height.

Furnitures

The furniture of ancient libraries played both a functional and spatial role. The arrangement of the furniture was designed for circulation and access to texts. In Roman libraries, the furniture was generally arranged along the walls, leaving the central space free for reading and scholarly activities [2].

7.6. Opening Hours

For building design, the Romans took climatic conditions into account to ensure greater visual comfort for readers. Opening hours indicated from the first to the sixth hour of the day, i.e., in the morning, from approximately 7:00 a.m. to noon. These times correspond to Vitruvius’s precepts. The rest of the time, access is closed by a gate or door, as evidenced by the archaeological remains of the Ulpian Library and the Library of Timgad. Hence, the simulation of the lighting environment is carried out based on reading hours, i.e., at 8:00 a.m., 10:00 a.m., and 12:00 p.m. on the solstices (21 June and 21 December) and the spring equinox (21 March).

7.7. Reading Books Related Behaviour

Regarding how to read the works, the library is not really highlighted apart from a few low reliefs. The Romans have often represented readers sitting at desks or on cushions and reading and studying in parchments or papyrus rolls. These different iconographic sources do not give us a clear idea of the current atmospheres within the ancient Roman libraries (Figure 16 and Figure 17).

8. Results

Even based on the users’ behavior inside this Roman heritage building, the following results characterize essentially and quantitatively the natural luminous environment of the main spatial component of the Timgad Roman library, that is, the reading room.

8.1. Numerical Simulation of the Main Reading Room

For the case of the main reading room of the library, the sequence has been divided into three (03) distinct zones (Sequences): (i) zone (a) comprising two large openings located on the same entrance wall, facing west, (ii) Zone (b) that is a partially enclosed space receiving diffused light from zone (a), and (iii) zone (c) where the reader stands to consult the documents in the wooden bookcase (Figure 18 and Figure 19). For the three zones, the luminance value is the outcome of this numerical simulation. Additionally, the luminance value is related to the vertical plane corresponding to the current user’s standing.

The numerical simulation outcomes, for the reading room’s lighting environment, reveal a non-uniform distribution of natural light depending on the dates and times of simulation. The simulation highlights contrasting lighting zones: (i) an intensively illuminated area (colored in red), and (ii) a dimly illuminated one (in blue). This difference suggests a marked influence of architectural elements, such as door and window heights. Such contrasting zones should create some problems in terms of spatially uniform visual comfort inside this main reading room.

Also, the simulation outcomes make known that the highest luminance values are specific to smooth, colored marble surfaces, in contrast to the lowest and related to rougher surfaces such as plaster (Figure 20, Figure 21 and Figure 22)

Such results clearly demonstrate that the surface characteristics of interior planes, namely floors, walls, and work surfaces (tables, shelves, furniture), significantly influence daylight distribution within the space. These planes can have reflective or absorbent properties, depending on the nature of the built materials used (polished marble, plaster, wood, stucco, etc.), which in turn determines the quality and quantity of diffused light.

Also, it must be noticed that certain architectural components are visually improved by natural light. This is the case of the main niche located at the back of the room, where the luminous environment is clearer than the other vertical planes constituting the whole visual field. Furthermore, the high amount of natural light in the center of the room gradually decreases on the reading room sides.

Consequently, it could be assumed that this light distribution results from several interdependent factors: (i) the spatial configuration of the room, (ii) the arrangement and size of the openings, (iii) the used building materials, and (iv) the function assigned to each area of the space.

8.1.1. Maximum Luminance

The findings indicate that the contrast at point (a) is particularly high, with a constant maximum value of 0.99 (99%) recorded throughout the day. This high level of contrast reflects a clear distinction between intensely illuminated areas and those partially or completely shaded, thus highlighting an accentuated play of dark and light in this specific area.

On the other hand, the contrast values are slightly lower at point (b) and range between 0.92 and 0.97. Although being less pronounced than those at point (a), these values are nevertheless high, indicating that illumination is still significant, but with a less pronounced transition between bright areas and the shaded ones, as shown in

Table 5. Level of contrast in different zones (cd/m²).

Date and Time	Sequence (a)			Sequence P (b)
	Lmax	Lmin	Contrast	Lmax	Lmin	Contrast
21 December at 10	519.1	3.6	0.99	85	2.9	0.96
21 March at 10	1142.6	4.8	0.99	88.3	3.0	0.96
21 June at 10	1557.1	5.4	0.99	91	3.4	0.96
21 December at 12	2272.6	5.8	0.99	90.6	3.9	0.95
21 March at 12	2609.9	5.6	0.99	92.4	2.6	0.97
21 June at 12	1732.2	9.8	0.99	93.5	3.7	0.96
21 December at 16	5878.4	6.0	0.99	96.7	6.8	0.92
21 March at 16	7124.0	7.1	0.99	99.5	4.7	0.95
21 June at 16	4702.3	7.2	0.99	98.7	3.7	0.96

.

According to the bar graph (Figure 23), we observe that, for sequence (a), the maximum luminance increases significantly in the middle of the day (12 p.m.) and reaches very high values in the afternoon, particularly on21 March at 4 p.m. with 7124.0 cd/m². The minimum luminance remains relatively stable, around 5 to 9 cd/m². The contrast remains consistently high (0.99), reflecting strong dynamic range.

For sequence P (b), the maximum luminance values are much lower than those of sequence (a), suggesting a configuration less exposed to natural light. The contrast is slightly lower (0.95–0.96), with a notable decrease at 4 p.m. in December (0.92).

8.1.2. Daylight Simulation of Natural Light: A Comparative Study Between Historical and Contemporary Results

Due to the lack of specific experimental data on illuminance levels in libraries, particularly in reading rooms, we referred to specialized authors in the field of historical public buildings and contemporary studies. Their findings were used for informational purposes to support the interpretation of our simulation results.

Unpredictably, thesesimulation-based values comply with the lighting-related contemporary ones recommended by M. Fontoynont [34] and C. Balocco and R. Calzolari [16], specialists in daylighting in historical buildings.

In fact, they insist on ensuring a good balance between diffuse and direct light in order to avoid dazzling while ensuring visual comfort. For these scientific researchers, the consultation of the works, in particular, when the characters are small, requires levels of lighting between 500 and 1000 lux, which can be obtained either with the ugly artificial light or near the windows.

Also, M. Fontoynont [34] insists on the fact that the daylight amount must be clearly attenuated in the storage areas in order to preserve the documents.

In Timgad’s library, although natural lighting is generally well exploited, its non-uniform distribution, in particular, the less enlightened side areas, suggests that these spaces are only reached in certain hours. In sequence (a), illumination levels consistently exceed recommended standards, ranging from 500 to 1000 Lux at all hours throughout the year. In contrast, sequence (b) remains within the reference thresholds for adequate lighting, with values maintained between 500 and 800 Lux. On the other hand, Sequence (c) corresponds to a dimly lit area, particularly during the month of December, when illumination decreases to 184.7 Lux (

Table 6. Illumination values in different zones (Lux).

Date and Time	Sequence (a)	Sequence (b)	Sequence (c)	Reference Values in (Lux)
December at 10	1719.9	272.5	184.7	500–1000 lux
March at 10	3023.0	500.5	354.9
June at10	4150.5	680	503.5
December at 12	3493.9	517.3	369.8
March at 12	5174.5	736.1	531.7
June at 12	6557.0	888.3	676.7

).

The gradual decrease in light intensity toward the sides allows us to formulate hypotheses about the functional organization of the room. Thus, it is possible that reading activities, requiring good lighting, took place mainly in the center of the room, where natural light is abundant. On the other hand, the less illuminated side areas were probably reserved for secondary functions such as consulting books, exchanges between readers or circulation.

9. Discussion

In the case of the Timgad library, the higher light intensity in the center of the reading room and the more marked shadow on the sides reflect an architectural intention aimed at prioritizing uses according to both quantity and quality of light.

As a result, the reading room of the Timgad library shows a significant variation in natural luminance depending on the time of day and the season, with very high maximum luminance levels in sequence (a), particularly at noon and in the afternoon. Conversely, sequence P (b), which is less exposed, has softer and more homogeneous lighting, providing more stable conditions, although sometimes insufficient for optimal reading without additional artificial lighting.

Adaptability of Ancient Lighting Devices: A Comparative Study Between the Library of Timgad and European Libraries

The library of Timgad shares some common architectural features with these Roman libraries, especially the number and arrangement of openings, often placed high above the wall niches used for storing scrolls. This configuration allowed the walls to remain free while ensuring uniform natural lighting. However, some aspects differ significantly. In Timgad, the openings are smaller in size and oriented toward the west, which is a notable feature. Given the intense sunlight typical of the North African climate, this west-facing orientation exposes the interior to direct solar radiation during the late afternoon hours. In order to reduce the light intensity, a portico was integrated into the West facade, serving as a solar filter. This portico creates a shaded area in front of the entrance, thereby reducing direct sunlight penetration into the reading room, especially during critical hours. This passive device helps moderate light intensity while maintaining ventilation and visual connection with the outside. The use of the portico, combined with less reflective materials such as limestone, illustrates an environmental adaptation strategy typical of Roman architecture in hot climates, fully exploited here to respond to the specific luminous conditions of the Timgad site.

10. Conclusions

This study approaches a historical ancient public building from an environmental point of view. In fact, the Rogatianus Roman library, in Timgad, has been investigated with a focus on its daylight inner conditions due to their importance as both functional and aesthetic design parameters.

Following a first stage consisting of a 3D model virtual restitution of this heritage public building, a numerical simulation has been carried out with respect to the users’ behaviors inside it. In such a way, the main reading room has been divided into a set of succeeding visual fields.

The simulation results clearly demonstrated that the distribution of light varies significantly across the different zones of the building, thereby influencing their functional suitability. The analysis revealed not only variations in luminance levels but also patterns of exposure related to spatial orientation and architectural configuration. This approach enabled the differentiation of spatial functions based on lighting intensity and distribution factors that had not been previously studied. It thus offered new insights into the relationship between architectural design and environmental performance, particularly in relation to space usage and thermal comfort.

This study highlights both the similarities and differences between the Timgad library and those located in Europe. Although these buildings share common architectural characteristics, such as the choice of materials and the number and arrangement of openings in height, they differ in their adaptation to the climatic conditions specific to each region. The Timgad library, located in a hot, semi-arid climate, is distinguished by adapted architectural strategies, including small openings and a portico.

Finally, the lack of wide information about the users’ behaviors inside Roman libraries is a substantial motivation to deeply investigate the iconographic as well as textual sources in order to clarify such confused knowledge, uncertainties and obscure aspects of the Roman libraries’ use and the daylight environment’s impact on it.